Influence of spatial structure on pathogen colonization and extinction: A test using an experimental metapopulation

Abstract

The Linum marginale-Melampsora lini plant-pathogen interaction has been studied extensively with regard to its epidemiology and population genetic structure (host resistance and pathogen virulence) in a natural metapopulation. In this study, this system was used in an experimental metapopulation approach to investigate explicitly how the distance (degree of isolation) between local population patches influences disease dynamics within a growing season, as well as the genetic structure of pathogen populations through stochastic colonization and extinction processes. The experimental design centred on four replicate sets of populations, within which patches were spaced at increasingly greater distances apart. Each patch consisted of an identical set of host and pathogen genotypes, with each pathogen genotype having the ability to attack only one of four host-resistance types. Over the 2 years of the experiment, the results showed clear 'boom-and-bust' epidemic patterns, with the strongest determinant of disease dynamics within a growing season being the identity of particular host-pathogen genotypic combinations. However, there were also significant effects of spatial structure, in that more isolated patches tended to exhibit lower levels of disease during epidemic peaks than patches that were close together. Extinction of pathogen genotypes from individual populations was positively related to the severity of disease during preceding epidemic peaks, but negatively related to the level of disease present at the final census prior to overwintering. The probability of recolonization of pathotypes into populations during the second growing season was most strongly related to the distance to the nearest neighbouring source population in which a given pathotype was present. Overall, these results highlight the importance of spatial scale in influencing the numerical and genetical dynamics of pathogen populations.